Large scale evaluation of local image featuredetectors on homography datasets
We present a large scale benchmark for the evaluation of local feature detectors. Our key innovation is the introduction of a new evaluation protocol which extends and improves the standard detection repeatability measure. The new protocol is better for assessment on a large number of images and red...
| Main Authors: | , |
|---|---|
| Format: | Conference item |
| Published: |
British Machine Vision Association
2018
|
| _version_ | 1797051615439486976 |
|---|---|
| author | Lenc, K Vedaldi, A |
| author_facet | Lenc, K Vedaldi, A |
| author_sort | Lenc, K |
| collection | OXFORD |
| description | We present a large scale benchmark for the evaluation of local feature detectors. Our key innovation is the introduction of a new evaluation protocol which extends and improves the standard detection repeatability measure. The new protocol is better for assessment on a large number of images and reduces the dependency of the results on unwanted distractors such as the number of detected features and the feature magnification factor. Additionally, our protocol provides a comprehensive assessment of the expected performance of detectors under several practical scenarios. Using images from the recently-introduced HPatches dataset, we evaluate a range of state-of-the-art local feature detectors on two main tasks: viewpoint and illumination invariant detection. Contrary to previous detector evaluations, our study contains an order of magnitude more image sequences, resulting in a quantitative evaluation significantly more robust to overfitting. We also show that traditional detectors are still very competitive when compared to recent deep-learning alternatives. |
| first_indexed | 2024-03-06T18:22:03Z |
| format | Conference item |
| id | oxford-uuid:06a121fc-cd06-425f-9f5f-d6b5c6d05742 |
| institution | University of Oxford |
| last_indexed | 2024-03-06T18:22:03Z |
| publishDate | 2018 |
| publisher | British Machine Vision Association |
| record_format | dspace |
| spelling | oxford-uuid:06a121fc-cd06-425f-9f5f-d6b5c6d057422022-03-26T09:03:31ZLarge scale evaluation of local image featuredetectors on homography datasetsConference itemhttp://purl.org/coar/resource_type/c_5794uuid:06a121fc-cd06-425f-9f5f-d6b5c6d05742Symplectic Elements at OxfordBritish Machine Vision Association2018Lenc, KVedaldi, AWe present a large scale benchmark for the evaluation of local feature detectors. Our key innovation is the introduction of a new evaluation protocol which extends and improves the standard detection repeatability measure. The new protocol is better for assessment on a large number of images and reduces the dependency of the results on unwanted distractors such as the number of detected features and the feature magnification factor. Additionally, our protocol provides a comprehensive assessment of the expected performance of detectors under several practical scenarios. Using images from the recently-introduced HPatches dataset, we evaluate a range of state-of-the-art local feature detectors on two main tasks: viewpoint and illumination invariant detection. Contrary to previous detector evaluations, our study contains an order of magnitude more image sequences, resulting in a quantitative evaluation significantly more robust to overfitting. We also show that traditional detectors are still very competitive when compared to recent deep-learning alternatives. |
| spellingShingle | Lenc, K Vedaldi, A Large scale evaluation of local image featuredetectors on homography datasets |
| title | Large scale evaluation of local image featuredetectors on homography datasets |
| title_full | Large scale evaluation of local image featuredetectors on homography datasets |
| title_fullStr | Large scale evaluation of local image featuredetectors on homography datasets |
| title_full_unstemmed | Large scale evaluation of local image featuredetectors on homography datasets |
| title_short | Large scale evaluation of local image featuredetectors on homography datasets |
| title_sort | large scale evaluation of local image featuredetectors on homography datasets |
| work_keys_str_mv | AT lenck largescaleevaluationoflocalimagefeaturedetectorsonhomographydatasets AT vedaldia largescaleevaluationoflocalimagefeaturedetectorsonhomographydatasets |